INDEX

Explanations

words related to observation or perception

oai_token-act-pair · gpt-3.5-turbo

instances of the word "seeing" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abin

-0.61

pro

-0.59

 Framework

-0.58

inals

-0.58

rog

-0.57

otom

-0.57

ors

-0.57

 Standards

-0.56

engine

-0.56

rete

-0.55

POSITIVE LOGITS

 seeing

3.27

 witnessing

2.18

 Seeing

1.98

 noticing

1.93

Seeing

1.91

seeing

1.85

 hearing

1.84

 encountering

1.69

 experiencing

1.62

 observing

1.57

Activations Density 0.016%

words related to observation or perception

instances of the word "seeing" in various contexts

No Comments

No Known Activations